microsoft / microsoft/markitdown

markitdown-mcp: server process dies on invalid_id / oversized input (Claude Code + Cursor)

Open
#2,146 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
186k
Forks
13.7k
Avg merge
1d 4h
Merged PRs (30d)
49

Description

While running through an MCP execution harness I built (usethrone/throne-ci), I hit a reproducible crash and wanted to share the full trace.
Running markitdown-mcp through my MCP harness (usethrone/throne-ci), the server process dies on error-handling probes:

text
invalid_id: died
oversized_input: died (stdin pipe closed)
recovery: died

Reproduces under both Claude Code and Cursor. connect/discover/schemas/reconnect all pass. Sealed record: [link]. Happy to re-run after a fix.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reproducing markitdown-mcp through the usethrone/throne-ci harness with the invalid_id and oversized_input probes, capturing the server's stderr and exit behavior. Compare these with the passing connect, discover, schemas, and reconnect probes; done means the server stays alive through both error cases and recovery works under Claude Code and Cursor.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
backend-api-design
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.